Abstract:Clinical prediction scores have been central to thrombosis and hemostasis practice for decades, providing transparent, interpretable, and validated approaches to risk assessment. Increasingly, artificial intelligence (AI)-based predictive models are being developed to improve risk stratification by integrating large, complex datasets and identifying patterns beyond conventional statistical models. However, improved predictive performance alone does not determine clinical value, and AI introduces new ethical, regulatory, and governance challenges. This narrative review compares traditional thrombosis risk scores and AI-based predictive models, focusing on methodological differences, ethical considerations, regulatory requirements, and the competencies clinicians require for responsible implementation. Traditional scores benefit from simplicity, transparency, and ease of clinical interpretation but have important limitations, including modest predictive performance, limited adaptability, potential bias, and uncertain evidence that score-guided management improves outcomes compared with clinical judgment alone. Many of the concerns currently associated with AI-based risk prediction, including bias, generalizability, validation, and monitoring, also exist within traditional prediction models but have historically received less regulatory attention. AI-based models offer potential advantages through dynamic risk prediction, integration of multimodal data, and improved predictive accuracy; however, they introduce additional challenges related to explainability, data governance, privacy, cybersecurity, accountability, reproducibility, and postdevelopment surveillance. AI should be viewed as an evolution of clinical decision support rather than a replacement for established approaches or clinical expertise. Safe implementation requires robust validation, continuous monitoring, human oversight, transparent governance frameworks, and clinician AI literacy.
Abstract:Blood, as a genetically determined phenotypic characteristic of red blood cells (RBCs), has traditionally been associated with transfusion safety and prevention of hemolytic reactions. However, recent data suggest that the ABO system may influence the predisposition to and the course of various pathological conditions, with potential implications for the development of comorbidities and clinical management of patients. Blood group antigens are expressed on RBCs, platelets, epithelial cells, and endothelial cells, and are involved in fundamental biological functions, such as cell recognition, intercellular adhesion, regulation of hemostasis, and immune response. Many epidemiological studies in adult populations demonstrate an association between specific blood groups of the ABO system and the Rh factor, and a differential risk of cardiovascular, metabolic, and infectious diseases, as well as the development of malignancies; however, in neonates, relevant studies are limited, and the impact of blood group on morbidity during the perinatal period remains unclear. The purpose of this narrative review is to investigate the association between neonatal blood group and the occurrence and severity of critical neonatal comorbidities, through an evaluation of the available clinical and epidemiological evidence, and to clarify the suspected pathogenetic mechanisms, with the aim of supporting future research efforts and improving the clinical care of this vulnerable population.
Abstract:The ONCO PE trial showed that 18-month rivaroxaban was superior to 6-month rivaroxaban for preventing recurrent venous thromboembolism (VTE) in patients with cancer-associated low-risk pulmonary embolism (PE). However, the influence of concomitant deep vein thrombosis (DVT) on this benefit was unclear. The current post-hoc subgroup analysis of the ONCO PE trial categorized patients into DVT subgroup (N = 104) and no DVT subgroup (N = 74) based on the presence or absence of a concomitant DVT at PE diagnosis. We compared the primary endpoint of 18-month recurrent VTE and the major secondary endpoint of 18-month major bleeding between the 18- and 6-month rivaroxaban groups in the DVT and no DVT subgroups. The cumulative 18-month incidence of recurrent VTE in the 18-month rivaroxaban group was numerically lower compared with that in the 6-month rivaroxaban group in the DVT subgroup (11.5 vs. 20.9%, log-rank p = 0.25) and was significantly lower in the no DVT subgroup (0.0 vs. 24.2%, log-rank p = 0.006), without significant interaction between treatment durations and concomitant DVT (Pinteraction = 0.99). There were no significant differences in the cumulative 18-month incidence of major bleeding between the 18- and 6-month rivaroxaban groups in both the DVT (4.6 vs. 5.8%, log-rank p = 0.70) and no DVT (13.4 vs. 5.8%, log-rank p = 0.29) subgroups, without significant interaction (Pinteraction = 0.34). The current exploratory analysis showed that extended rivaroxaban treatment might have a potential benefit for patients with cancer-associated low-risk PE in preventing recurrent VTE, irrespective of concomitant DVT status, without a significantly increased risk of major bleeding.
Abstract Neonatologists spend considerable effort with the aim of preventing neonatal bleeding. Traditionally, platelet failure has been widely considered a crucial determinant of neonatal hemorrhage. However, recent research challenges this view and suggests that platelet hyporeactivity may be part of a carefully balanced hemostatic system. This review explores the broader process of primary hemostasis in both term and preterm infants, beyond platelet function alone, integrating evidence from in vitro, laboratory, and point-of-care studies. Overall, studies raise concerns about the possible contribution of platelet dysfunction to an increased bleeding risk, especially in preterm infants. The clinical relevance of these findings remains uncertain, owing to methodological heterogeneity and the limited correlation with meaningful outcomes. Although platelet hyporeactivity has been largely attributed to a developmental impairment, it might also depend on prevailing up-regulated inhibitory pathways. Nevertheless, functional differences in plasma proteins provide enhanced adhesive capacities, supporting platelet-subendothelium interaction, further facilitated by higher hematocrit levels. Molecular explanations for platelet singularities are being uncovered thanks to omics studies. Advanced technologies such as microfluidics are enabling the study of cross-talking endothelium and platelets. Yet, the details of the interactions between platelets, erythrocytes, and white blood cells remain to be elucidated in the setting of developmental hemostasis. Emerging evidence on distinct pathways driving a lower pro-inflammatory tendency of the neonatal milieu might explain the poorer outcomes of those preterm infants receiving multiple transfusions of adult blood products. Awareness of the unique equilibrium of neonatal primary hemostasis calls for further research into the impact of neonatal therapies on hemostatic balance.
Abstract:The optimal timing of pharmacological thromboprophylaxis in traumatic brain injury (TBI) remains uncertain due to competing risks of thrombosis and bleeding. This study evaluated whether early heparin initiation (≤72 hours) reduces vascular occlusive events (VOEs), mortality, and hospital stay without increasing bleeding. We compared early versus late (>72 hours) heparin use via propensity score matching (PSM), subgroup analyses, and multivariable logistic regression. Outcomes included VOEs (deep vein thrombosis, pulmonary embolism, myocardial infarction, and ischemic stroke), bleeding, 30- and 180-day mortality, and length of stay. After PSM, late heparin was associated with higher VOEs (25.16 vs. 12.24%, p < 0.001), 30-day mortality (20 vs. 11.89%, p = 0.03), 180-day mortality (32.26 vs. 19.58%, p = 0.004), and prolonged hospitalization (median: 15.7 vs. 9.7 days, p < 0.001), without increased bleeding risk (16.77 vs. 22.03%, p = 0.24). Subgroup analyses revealed late heparin as an independent risk factor for VOEs in mild TBI (odds ratio: 2.62, 95% confidence interval: 1.54-4.45; p < 0.001) regardless of fracture status, coagulation profile, or hemorrhagic classification, with nonsignificant associations in nonhemorrhagic and moderate-to-severe TBI subgroups. Multivariable analysis identified late heparin, higher comorbidity index, elevated platelet count, and cranial surgery as independent predictors of VOEs, with late heparin also independently associated with increased mortality. These findings suggest early heparin initiation reduces thromboembolic complications, mortality, and hospital stay without markedly increasing bleeding risk, supporting early anticoagulation in selected patients following individualized risk-benefit assessment.
Abstract Plasminogen activator inhibitor (PAI-1) is a member of the serine protease inhibitor family critical in regulating fibrinolysis and is associated with diverse pathologies including cardiovascular disease, metabolic disorders, cancer, aging, and many others. Despite its clinical significance and extensive research, there are as of yet still no FDA- or EMA-approved therapeutics that specifically target PAI-1. This review examines therapeutic strategies including small molecules, RNA aptamers, monoclonal antibodies, and repurposed pharmaceuticals. Notable challenges in the development of these therapies include the interplay of PAI-1 with tissue-specific mediators such as vitronectin and the multi-dimensional, tissue-specific effects of PAI-1 modulation. Promising approaches will require tissue- and temporal expression-based interventions. Further characterization of PAI-1's structural dynamics and its role in various disease processes will shape its potential as a therapeutic target across broad fields in medicine.
Abstract:Sex-related differences in outcomes after major orthopaedic surgery remain under-investigated, and the prognostic relevance of a history of venous thromboembolism (VTE) is unclear. The ORTHO-START registry is a prospective, multicenter study including 1,077 consecutive adults undergoing major orthopaedic surgery (elective hip/knee arthroplasty or urgent hip fracture repair) across six Italian centers between July 2018 and March 2022. Baseline characteristics, perioperative complications, and antithrombotic therapy were collected. The primary outcome was all-cause mortality at 24 months; secondary outcomes were thromboembolic and bleeding events. Multivariable logistic regression identified predictors of mortality. Women represented 72.3% (n = 775) of the cohort and were older than men (80.6 ± 10.0 vs. 76.2 ± 13.0 years, p < 0.001). Urgent procedures comprised 65.6% (n = 706) of surgeries, predominantly involving women (n = 542). At 24 months, mortality was higher in men (31 vs. 18%, p = 0.018). Independent predictors of mortality included male sex (odds ratio [OR] = 2.6, 95% confidence interval [CI]: 1.8-4.0), older age, lower body mass index, urgent surgery (OR = 7.2, 95% CI: 4.1-12.4), and prior VTE (OR = 5.7, 95% CI: 1.4-23.9). Causes of death were known for 160/210 (76.2%): neurological complications were most frequent (25%), followed by cardiovascular disease (21.3%) and immobility-related causes (14.4%); pulmonary embolism accounted for 1.3% of deaths. Thrombotic and hemorrhagic events occurred in both sexes without significant differences. Male sex and a history of VTE are independent predictors of long-term mortality following major orthopaedic surgery. Incorporating these factors into preoperative risk assessment may improve management and postoperative outcomes.
Abstract von Willebrand disease (VWD) is the most commonly inherited bleeding disorder, with a prevalence surpassing hemophilia A. Unfortunately, VWD may be variously underdiagnosed, overdiagnosed, or misdiagnosed, depending on the expertise of the managing clinician and testing laboratories. Diagnostic challenges are due to the heterogeneity of VWD and the complexities surrounding laboratory assessment, including reactive influences on VWF levels. At least six types of VWD can be identified, with these classified according to the functional defect and/or level of deficiency in the plasma protein von Willebrand factor (VWF). The VWF protein has several functions, most of which can be assessed by laboratory testing; however, this increases the diagnostic complexity, and clinicians/laboratory staff may not understand the differences in these tests and what they assess. Thus, a battery of laboratory tests is required to enable an effective diagnosis or exclusion of VWD, as well as its type classification. VWD was first identified in a young female patient by Erik von Willebrand in 1924, with a seminal publication on his findings appearing in the literature in 1926. This year, 2026, represents 100 years of VWD. We review some of the history of VWD, as well as outlining the contemporary diagnostic pathway for VWD, assisted by several illustrative case examples.
Abstract:This study aimed to analyze the interim efficacy of different anticoagulation regimens for patients with long-term inferior vena cava filter (IVCF) implantation. From August 2021 to August 2024, a retrospective analysis of patients who underwent long-term IVCF implantation in multiple centers was performed. Group A (46 cases) did not receive anticoagulation, Group B (66 cases) received extended secondary prophylaxis dose anticoagulation, and Group C (63 cases) received extended therapeutic dose anticoagulation, with 12 months follow-up. The primary endpoint was lower extremity deep vein thrombosis (LEDVT) recurrence rate. The secondary endpoints were IVCF thrombosis (IVCFT) incidence, bleeding-related adverse events incidence, all-cause mortality, and venous clinical severity score (VCSS). LEDVT recurrence rate and IVCFT incidence in Group A (37.0 and 19.6%) were higher than that in Group B (13.6 and 6.1%) and Group C (15.9 and 4.8%). Bleeding-related adverse events incidence in Group C (17.5%) was higher than that in Group B (4.5%) and Group A (2.2%). There was no significant difference in all-cause mortality among the three groups. VCSS at 12 months in Group A was higher than that in Group B and Group C. In the interim follow-up of the extended treatment phase in patients with long-term IVCF implantation and rivaroxaban anticoagulation may reduce LEDVT recurrence and IVCFT occurrence compared with non-anticoagulation, and alleviate patients' clinical symptoms. The extended secondary prophylaxis dose may exhibit comparable efficacy to the extended therapeutic dose, while potentially reducing the bleeding risk.
Abstract:Treatment of catheter-related thrombosis (CRT) in acute leukemia patients is challenging due to concomitant thrombocytopenia. Although modified-dose anticoagulation is frequently used in this setting, several vital outcomes remain unclear. The current study evaluated clinical outcomes of a platelet-count-guided low molecular weight heparin (LMWH) regimen in these patients. Data on all newly diagnosed acute leukemia patients, who developed CRT, were retrieved from the institutional database. Upon CRT diagnosis, patients received either a full or modified dose of LMWH, based on their platelet counts. Outcomes included recurrent venous thromboembolism (VTE), hemorrhagic events. Competing risk analyses were performed, with death considered a competing risk. All-cause mortality and blood product utilization (per patient; pre- and postanticoagulation commencement) were compared. A total of 193 acute leukemia patients diagnosed with CRT between November 2004 and May 2022 were included. Eighty-seven patients (45%) received full-dose LMWH. Modified-dose LMWH treatment was associated with hazard ratio (HR) of 0.642 (95% confidence interval [CI]: 0.268-1.540; p = 0.320) in terms of CRT recurrence and HR of 0.975 (95% CI: 0.439-2.170; p = 0.950) in terms of hemorrhage. Mortality was not significantly higher in the modified-dose group (HR: 1.47, 95% CI: 0.93-2.32; p = 0.092). Notably, platelet requirement did not differ before and after the CRT diagnosis, irrespective of LMWH doses used (full-dose group: median pre-CRT 0.09 [interquartile range, IQR: 0-0.27] vs. post-CRT 0.05 [IQR: 0.01-0.14] units/day, p = 0.059; modified-dose group: median pre-CRT 0.03 [IQR: 0-0.24] vs. post-CRT 0.07 [IQR: 0.01-0.18] units/day, p = 0.720). In acute leukemia patients with CRT, a platelet-count-guided LMWH regimen was not associated with higher rates of recurrent VTE, hemorrhagic complications, mortality, or platelet transfusion requirements.
Abstract:Platelets play a fundamental role in hemostasis; however, their phenotype, reactivity, and function vary across the pediatric age spectrum. While developmental differences in plasma coagulation proteins have been well-characterized in children, differences in platelet biology remain poorly understood, with most studies focused on neonates. Consequently, the understanding of normal platelet phenotype and function in healthy children remains limited, making it difficult to interpret platelet changes in the context of disease. Platelet phenotype evolves from birth through adolescence, with ongoing changes in receptor expression and granule content. Functional responses also change with age, with neonatal platelets being hyporeactive compared with adult platelets, whereas platelets from older children may demonstrate increased responsiveness to common agonists. These developmental differences are increasingly recognized as clinically relevant across a range of pediatric disease states. However, key knowledge gaps remain in age-specific reference ranges and the translation of platelet biology into clinical practice. This review summarizes the current evidence on developmental differences in platelet production, phenotype, and reactivity from birth to adolescence and examines their clinical significance in pediatric disease states.
Abstract:Haemostasis reflects a balance between procoagulant and anticoagulant forces. A relative excess of procoagulant forces and/or a relative lack of anticoagulant forces can predispose individuals to an increased risk of thrombosis. Conversely, a relative lack of procoagulant forces and/or a relative excess of anticoagulant forces can predispose individuals to an increased risk of bleeding. It is possible in several clinical situations to mitigate bleeding risk using various approaches, including provision of procoagulant material; conversely, thrombosis risk can be mitigated by using various clinical anticoagulants. In some cases, such therapies require laboratory monitoring or measurements; in particular, therapy using anticoagulants may require laboratory measuring or monitoring to ensure adequate anticoagulation without over- or under-coagulating patients, and thereby reducing any treatment associated risk of bleeding or thrombosis. In this narrative review, we detail the various anticoagulants under current clinical use, as well as the tests or assays used to measure anticoagulant levels, or to monitor anticoagulant therapy.
Abstract:The human coagulation system has resulted from thousands of years of evolutionary pressure, finely balancing the competing risks of lethal hemorrhage and catastrophic thrombosis. There are several examples of the close interaction between the two opposing sides of hemostasis. A paradigmatic case is offered by the factor V Leiden mutation. Emerging approximately 22,000 years ago, this gain-of-function variant is now expressed in approximately 5% of Caucasians. While it now significantly elevates the risk of venous thromboembolism, it historically provided crucial evolutionary advantages, including reduced peripartum blood loss, enhanced iron stores, improved fertility, and increased embryo implantation rates. Conversely, congenital hypocoagulable states, such as hemophilia, demonstrate the other side of the coin. With modern therapeutic advances now extend the life expectancy of individuals with hemophilia, clinicians are increasingly managing age-related comorbidities. While experimental models suggest a chr onic hypocoagulable state might mitigate the risk of occlusive cardiovascular events or suppress cancer metastasis, clinical and epidemiological data remain conflicting. Mild factor deficiencies may offer some protection against ischemic heart disease, but overall cardiovascular and oncological risk profiles in patients with hemophilia increasingly mirror those of the aging general population. This review explores the genetic and environmental forces that shape this homeostatic equilibrium, illustrating how historical survival mechanisms influence modern disease patterns.
Abstract:The management of hemophilia A has been transformed by the introduction of extended half-life factor VIII (EHL-FVIII) products. Similarly, the development of extended half-life factor IX (EHL-FIX) products has also impacted the treatment of hemophilia B, improving treatment adherence and reducing bleeding episodes. These bioengineered molecules, created through Fc-fusion, PEGylation, single-chain technologies, or albumin conjugation (particularly for FIX), offer patients reduced infusion frequency while maintaining effective bleed protection. However, the structural modifications of both EHL-FVIII and EHL-FIX pose significant challenges for accurate laboratory monitoring. Traditional one-stage and chromogenic substrate assays often yield discrepant results, varying by reagent, analytical platforms, and product type, complicating clinical decision-making and pharmacokinetic assessments. The thrombin generation assay (TGA) has emerged as a powerful global hemostasis tool that provides a functional readout of a patient's overall coagulation potential. By measuring the enzymatic capacity to generate thrombin over time, TGA offers a more integrated view of hemostatic efficacy than conventional factor activity tests. Although promising, TGA remains limited by lack of standardization, interplatform variability, and absence of validated clinical thresholds. Current data support its use as a complementary tool rather than a replacement for conventional assays. Future prospective studies integrating laboratory and clinical endpoints are required to define its role in precision hemophilia care and to determine whether TGA can better reflect bleeding phenotype and guide individualized therapy. Ultimately, the integration of functional assays into clinical practice will depend on harmonization efforts and demonstration of added clinical value in real-world settings and trials.
Abstract:Low molecular weight heparin (LMWH) remains a widely utilized parenteral anticoagulant in hospital and outpatient settings for thromboprophylaxis and treatment of venous thromboembolism. While LMWH typically does not require routine monitoring due to predictable pharmacokinetics, anti-factor Xa (anti-Xa) monitoring is indicated in specific populations, including those with renal impairment, extremes of body weight, pregnancy, or unexplained treatment failure. LMWH monitoring relies exclusively on chromogenic anti-Xa assays, as LMWH's predominant anti-Xa activity with minimal anti-IIa effect renders activated partial thromboplastin time insensitive. Therapeutic targets are context-dependent: peak anti-Xa levels of 0.5 to 1.2 U/mL for twice-daily therapeutic dosing or 1.0 to 2.0 U/mL for once-daily dosing, measured 3- to 5-hour postdose. We report external quality assessment findings for anti-Xa testing of LMWH from the past 6 years (2020-2025 inclusive) using RCPAQAP (Royal College of Pathologists of Australasia Quality Assurance Program) data, an international program with approximately 150 laboratory enrolments. Four samples are assessed annually at varied LMWH concentrations (0, 0.2-0.3, 0.5-0.6, 0.8-1.0 U/mL). Good reproducibility was demonstrated for duplicate samples across surveys. Coefficient of variation data revealed moderate variability (10-25%) for detectable LMWH levels, lower than unfractionated heparin (UFH) monitoring variability. Method-related differences were observed, with Siemens Innovance anti-Xa assays yielding slightly higher values than Werfen HemosIL or Stago STA assays. Participant interpretations of LMWH levels (below/within/above therapeutic range) showed greater variability than UFH due to multiple valid therapeutic ranges for LMWH depending on dosing context. Overall, laboratories demonstrated acceptable analytical and interpretative performance for LMWH monitoring, which should provide some reassurance to both patients and their clinical carers.
Abstract:Extended anticoagulation is recommended in subjects with cancer-associated thrombosis (CT). The API-CT trial showed that reduced-dose apixaban is noninferior to full dose for preventing recurrent venous thromboembolism (VTE), but the generalizability of these findings to real-world populations remains to be determined. We conducted a multicenter retrospective cohort study including patients with CT receiving extended direct oral anticoagulants (DOACs) at full or reduced dose, after at least 6 months of anticoagulation. The primary efficacy outcome was recurrent VTE. The primary safety outcome was major bleeding (MB) and clinically relevant nonmajor bleeding (CRNMB). A total of 603 patients were included, of whom 381 (63.2%) received reduced-dose and 222 (36.8%) full-dose DOACs. During a median follow-up of 335 days, recurrent VTE occurred in 3.9% of patients in the reduced-dose and 2.7% in the full-dose group (p = 0.4). MB occurred in 1.0 and 0.9% (p = 0.9) and CRNMB in 4.7 and 5.4% (p = 0.7), respectively. In adjusted analyses, reduced-dose DOACs were associated with a nonsignificant increase in VTE recurrence (hazard ratio: 1.4, 95% confidence interval: 0.5-3.7), with no differences in bleeding outcomes. In this multicenter real-world cohort, reduced-dose DOACs were associated with low rates of recurrent VTE and bleeding, comparable to those observed with full-dose regimens. These findings suggest that reduced-dose anticoagulation may be a feasible option in selected patients with CT requiring extended treatment, supporting individualized decision-making. Further studies are needed to better define optimal dose selection in this setting.
